Nonlinear relations between achievement and academic self-concepts in elementary and secondary school: An integrative data analysis across 13 countries.

Abstract

It is well-documented that academic achievement is associated with students' self-perceptions of their academic abilities, that is, their academic self-concepts. However, low-achieving students may apply self-protective strategies to maintain a favorable academic self-concept when evaluating their academic abilities. Consequently, the relation between achievement and academic selfconcept might not be linear across the entire achievement continuum. Capitalizing on representative data from three large-scale assessments (i.e., TIMSS, PIRLS, PISA; N = 470,804), we conducted an integrative data analysis to address nonlinear trends in the relations between achievement and the corresponding self-concepts in mathematics and the verbal domain across 13 countries and two age groups (i.e., elementary and secondary school students). Polynomial and interrupted regression analyses showed nonlinear relations in secondary school students, demonstrating that the relations between achievement and the corresponding self-concepts were weaker for lower achieving students than for higher achieving students. Nonlinear effects were also present in younger students, but the pattern of results was rather heterogeneous. We discuss implications for theory as well as for the assessment and interpretation of self-concept.

School innovativeness is associated with enhanced teacher collaboration, innovative classroom practices, and job satisfaction.

“…Second, ensuring that latent variables are invariant across countries and levels at the same time is, to our best knowledge, currently only possible via fixing the parameters of the measurement models for each country data set. This fixed-parameter approach may deteriorate model fit and introduce some bias to the structural parameters (Scherer & Nilsen 2016), but because the state-of-research indicates that this bias may be considerably small (Devlieger & Rosseel, 2019), we consider the SAM approach a promising alternative for analyzing large-scale educational data (Keller et al, 2020).…”

“…We followed the Cheung and Jak (2016) approach to quantify and explain between-country variation, that is, the SAM approach. In this approach, the large-scale data are first split into different samples—in our study, the countries (Keller et al, 2020). Second, the analytic models of interest are specified, estimated, and evaluated for each sample—in our study, the multilevel structural equation models with the fixed parameters in the measurement models.…”

How much do students’ scores in PISA reflect general intelligence and how much do they reflect specific abilities?

“…The studies cited above on the latent structure of LSAs clearly identified general ability factors in addition to domain-specific abilities. However, these findings did not change practices within the psychometric community involved in the construction and scaling of LSAs, nor in the education academic research and policy communities that analyze and interpret LSA data for research and to inform policy making (e.g., Borgonovi & Pokropek, 2019;Deng & Gopinathan, 2016;Jakubowski & Pokropek, 2015;Keller et al, 2020). Measurement frameworks in education do not consider the relevance of g. Brunner (2008) notes that analyses of students' characteristics and domain-specific ability scores reported in LSAs, meta-analyses, and reviews have relied almost entirely on the standard model of domain-specific abilities.…”

Meta-Analyzing Individual Participant Data From Studies With Complex Survey Designs: A Tutorial on Using the Two-Stage Approach for Data From Educational Large-Scale Assessments

“…Such research opportunities involve in-depth meta-analyses of policy-relevant subgroups, for example, to provide reliable empirical evidence of the consistency and generalizability of the magnitude of gender differences among top-performing students (i.e., the top 5%) in mathematics (Keller et al, 2021). Likewise, IPD meta-analyses of ELSAs allow researchers to examine the generalizability of (novel) theoretical propositions across countries, for example, to test predictions about nonlinear relationships between achievement and academic self-concepts (Keller et al, 2020), relationships between innovative school environments and teaching practices (Blömeke et al, 2021), relationships between epistemic beliefs and educational outcomes in science (Guo et al, 2021), year-in-school effects on academic selfconcept (Marsh, 2016), or, as we illustrate in the present paper, how gender differences in students' achievement are moderated by their SES (see Else-Quest & Hyde, 2016).…”

“…However, meta-analytic models have only rarely been applied to results from ELSAs in general (e.g., Else-Quest et al, 2010; and to IPD from ELSAs in particular (e.g., Blömeke et al, 2021;Gray et al, 2019;Keller et al, 2020Keller et al, , 2021Nowell & Hedges, 1998). 1 Why?…”